Ambient Diffusion Omni (Ambient-o) is a framework for using low-quality, synthetic, and out-of-distribution images to improve the quality of diffusion models. Unlike traditional approaches that rely on highly curated datasets, Ambient-o extracts valuable signal from all available images during training, including data typically discarded as "low-quality."
This model card is for a model trained on ImageNet. Unlike normal training, for this run, low-quality images from ImageNet were only used to train only for certain diffusion times, but not other.
At high diffusion times, the model leverages the theoretical insight that noise contracts distributional differences, reducing mismatch between high-quality target distribution and mixed-quality training data. This creates a beneficial bias-variance trade-off where low-quality samples increase sample size and reduce estimator variance.
At low diffusion times, the model exploits locality properties of natural images, using small image crops that allow borrowing high-frequency details from out-of-distribution or synthetic images when their marginal distributions match the target data.
1@article{daras2025ambient,
2 title={Ambient Diffusion Omni: Training Good Models with Bad Data},
3 author={Daras, Giannis and Rodriguez-Munoz, Adrian and Klivans, Adam and Torralba, Antonio and Daskalakis, Constantinos},
4 journal={arXiv preprint},
5 year={2025},
6}
